Abstract
Twenty men and 20 women with children under 3 years of age, and 20 men and 20 women without children were asked to select the most suitable utterance for a series of drawings representing different contexts of mother/father-infant interaction. Data were analyzed in three sections related to different hypotheses: In Phase 1, informational content and syntactic construction of sentences were selected in strict relationship with context by all speakers (“closeness to context” rule); in Phase 2, different speech acts were selected by different speakers within the same context (analysis of illocutionary force and canonicality); in Phase 3, the same contexts were linked to different interactional dimensions for the four groups, which formed different representations of the same interactional scene. Results are discussed in an attempt to integrate specific rules of adult-infant interaction and more general aspects of discourse production.